4 replies to this topic

[p.bambury]

Hello, 

 

I am working on a cross-over dataset: all my patients received three different treatments, with a washout period. The time in the database starts therefore at t=0 for each of the three periods.

I thought I should use the column RESET for the PK analysis, to inform Phoenix that it is the same individual (with the same etas), with RESET=1 at pre-dose time. However, when I tried, Phoenix was unable to run because it doesn't recognize my covariates ... Is there a problem with covariates when using RESET column? Is there an other way to estimate PK parameters from a cross-over?

Thank you.

[smouksassi1]

Hi we need more details to help.

REset works for cross over and let time start again from zero covariate can have different values across periods or can be time varying within a period

you need to tell it what flag/value start a period with min max traditionally EVID 3/4

[p.bambury]

Thank you for your answer.

I have a column of time with two periods, each starting at t=0. Patients receive the dose at t=0. I also have a covariate which is constant within a period but varies across period. My issue is that when I try to run my model, I have this execution error "Covariate not set". But it is actually set ... If I don't use the RESET column but set treatment as ID instead, the model runs... So I don't understand why the covariate is not seen when I use the RESET column.
Thank you for your help!

Hi we need more details to help.

REset works for cross over and let time start again from zero covariate can have different values across periods or can be time varying within a period

you need to tell it what flag/value start a period with min max traditionally EVID 3/4

[smouksassi1]

as mentioned we need data or project to help so we can see what is going on not sure what do you mean treatment as ID how many ID/treatment groups you have do you have multiple treatments in the same ID how does the EVID column look like in your data you need to reset once at the start of the period

[Simon Davis]

Hi Pauline, it is possible this is an known issue (QC17731).

 

NLME does not support a separate dosing panel when reset+covariates is in use. The workaround is to merge doses into the main panel.  Please see if that helps.

 

  Simon.